Connectivity-Based Brain Parcellation for Parkinson's Disease
Yu Li, Aiping Liu, Liangyong Li, WU Yun-hu, Martin J. McKeown, Xun Chen, Feng Wu
Abstract
Connectivity-based parcellation (CBP) studies for exploring cerebral topographic organization have emerged rapidly, likely due to the joint developments of non-invasive imaging technologies and advances in computing science. CBP studies have extended our understanding of human brain development and many brain-related disorders such as Parkinson's Disease (PD), and have provided promising approaches to guide electrode placement during the planning of deep brain stimulation (DBS) surgery. This work reviews prevalent CBP methods, summarizing the methodological advantages and limitations of each. As PD is the second most common neurodegenerative disorder, we particularly focus on data-driven parcellation studies in this disease, providing researchers with a comprehensive overview of PD-specific atlases and their applications. We show that, while many advances have been achieved, heterogeneity in the PD population still provides an ongoing challenge to find a robust consensus on regional representation. Although some parcellation-driven studies exhibit encouraging achievements, these PD-specific parcellations are still limited and most approaches depend on a single modality. We discuss the future directions of parcellation-driven PD exploration and surgical planning, with the aim to inspire future investigation into connectivity-based parcellation for PD.